CARBOXYLIC Preparation of Peptides Using Mixed Carboxylic Acid

measurable autocatalytic period up to pH 7.9. Joslyn and Branch" found no autocatalytic period for catechol. The inverse hydrogen ion relationship sho...
1 downloads 0 Views 404KB Size
Dec., 1951

PEPTIDES FROM MIXEDCARBOXYLIC ACID ANHYDRIDES

is solvent or hydrogen peroxide and ks is the rate constant for the reaction between quinone (T) and X. I n the case of chlorogenic acid ka is probably small, since the quinone formed is more stabilized by resonance than o-benzoquinone. I n addition, one of the two reactive positions (the 4,5-carbons) is substituted and is unavailable. This substitution must offer some steric hindrance to further reaction a t carbon atom 5. The autoxidation of chlorogenic acid also differs from that found for catechol by the occurrence of a measurable autocatalytic period up to pH 7.9. Joslyn and Branch" found no autocatalytic period for catechol. The inverse hydrogen ion relationship shows that the catechol ion with one hydroxyl group ionized is very probably the reactive species. At PH values greater than 8.74 the rate constants increase as Joslyn and Branch" also found for catechol at high pH values. Their explanation that this is due to the oxidation of the ion with both hydroxyl groups ionized may also be true here. However, i t is also possible that the high rate constants a t the higher pH values may be due to an inaccuracy in the extrapolation of rate to zero time. I n some runs, which were carried up to 30% reaction, the oxygen uptake curve (e.g., B of Fig. 1) appeared to be linear throughout. This linearity would seem to indicate a zero-order relationship for chlorogenic acid during the reaction, which might be expected if the products absorbed oxygen a t a comparable rate to chlorogenic acid. I n order to determine what the course of the oxygen uptake would be during the reaction if the

[CONTRIBUTION

FROM THE

5553

products did not absorb oxygen, equation 1 was integrated. Integration of equation 1 gives the equation Vol. 02 = 22400C0[l - exp( -kP0J(H+)22400)]

(4)

where COis the initial concentration of chlorogenic acid and C is the concentration of chlorogenic acid a t any time during the reaction. The assumptions were made that (a) there is a 1 : l stoichiometry between oxygen and chlorogenic acid, (b) PH and Po2 remain constant during the reaction, and (c) chlorogenic acid disappears by the first order rate law C = COexp. ( - k Po,t/(H+)22400) Using an average value of b for experiments 1-17 and the initial conditions of curve B (Fig. l ) , curve D (Fig. 1) was calculated from equation 2. As curve B does not differ significantly from curve D, it is not possible to draw any conclusions concerning the order of chlorogenic acid during the reaction or whether or not the products of the chlorogenic acid oxidation themselves absorb oxygen. Further research is being undertaken on this point. Since the k we found is a product of kl, kq and K , the energy of activation contains contributions from three sources and is therefore difficult to compare with the corresponding value found by Joslyn and Branch for catechol, which also includes contributions from similar terms. Acknowledgment.-We wish t o thank Dr. Benjamin Makower for many stimulating discussions during this work. ALBANY6, CALIFORNIA

RECEIVED MAY8, 1951

CHEMOTHERAPY DIVISION, STAMFORD RESEARCH LABORATORIES, AMERICANCYANAMID COMPANY]

Preparation of Peptides Using Mixed Carboxylic Acid Anhydrides BY JAMES R. VAUGHAN, JR., AND RUTHL. OsilTo Mixed anhydrides of N-substituted aminoacids with a- or B-branched chain, low molecular weight aliphatic acids are formed under anhydrous conditions at low temperature and react with aminoacid esters to give good yields of the corresponding peptide esters. Anhydrides with isovaleric acid have been found particularly advantageous and have been used t o prepare a series of peptides. The theoretical background for the work is discussed.

The use of mixed carboxylic acid anhydrides for the synthesis of peptides was reported recently by Wieland and Sehring,' who found that anhydrides of benzoic or acetic acid with N-substituted aminoacids react readily with aqueous solutions of a salt or an ester of a second aminoacid to give the corresponding peptide derivatives. No studies of the peptide reaction under anhydrous conditions were reported. This was of interest in view of the recent work of Emery and Gold2on the reactivities of mixed anhydrides in which they stress not only the importance of steric hindrance and charge-distribution in the molecular anhydride, but also the importance of the polarity of the solvent on the acylation ratio obtained bhen the (1) T. Wieland and R. Sehring, Ann., 519, 122 (1950). (2) A. R. Emery and V. Gold, J. Chcm. Soc., 1443, 1447, 1455 (1950).

anhydride reacts with an amine. According to these authors, under anhydrous conditions nucleophilic attack in the molecular anhydride will occur a t that carbonyl carbon atom which has the lowest electron density and, other things being equat, which is less hindered sterically. Under aqueous conditions, however, these considerations become invalid, possibly due to ionization of the anhydride, and a marked change or even a reversal of the acylation ratio is observed. Predictions concerning the reaction in polar solvents, therefore, are uncertain. Under anhydrous conditions, however, the anhydride forming acid, the acylation reaction of which is to be suppressed, should be one in which the combined operation of an electronic and a steric effect would lessen the probability of nucleophilic attack a t this part of the anhydride molecule and favor substitution a t the

5554

JAMES

K. VAUGHAN, JR.,

AND

RUTHL. OSATO

VOl. 73

relatively less hindered and relatively more elestrophilic aminowid carbonyl carbon. In the present work, mixed anhydrides were formed between caxbobenzoxyglycine and a series of both aliphatic and aromatic acids by reaction the respective acid chlorides with the triethylamine salt of carbobenzoxyglycine under anhydrous, standardized conditions. The anhydrides were then treated with aniline and the yields of carbobenzoxyglycinanilide taken as a measure of the acylation ratio. The over-all reaction may be formulated as

and inductive effect, gave less favorable results and a less pure product. Attempts to demonstrate a n effect attributable to increased nterie hindrance in these compounds by introducing various substituents into the benzene ring gave inconclusive results. When trimethylaeetic, diethylacetic and isovaleric carbobenzoxyglycine anhydrides were further compared under anhydrous conditions in the preparation of ethyl carbobenzoxyglyeyl-DLphenylalaninate, the yield of peptide ester was 81, 84 and 86%) respectively. Under conditions using wet solvents, however, the yields of peptide ester were reduced to %,67 and 55y0. \ CH20C0NHCH2COOHS ( C 2 H , ) > + KCOCl -+ Since the isovaleric mixed anhydride (R = (CHa)&HCHs-) seemed eminently satidactory for [/>CH~OCO?\THCH~COOCOR] + (C~H,),S.HCI the purpose intended, isovaleryl chloride was chosen as a suitable reagent and used in the preparation of the peptide esters listed in Table 11. As additional examples were obtained using the reaction, it was found to be more satisfactory than originally expected. The conditions of low tem~ ~ C H ~ O C O S H C H ~ C O KCOOH I H ~ + perature under which the reaction is carried out prevent excessive decomposition or by-product and the results are summarized in Table I. formation and lead to products of high initial purity. -It higher temperatures the yields are TABLE I lower and the produsts are less pure. Also, PREPARATION OF CARBOBENZOXYCLYCINA~ILIDE B Y THE racemization does not seem to occur under these MIXEDCARBOXYLIC ACIDANHYDRIDE PROCECIJRE conditions, since several optically-active peptides Acid chloride Yield, Vo M p (COT)~ prepared by this and by other methods have identical rotations. The expected difficulty of sepaAcetyl 36 144-145 rating the desired peptide product from the product Ilimethylace tyl 65 146-147 obtained when the isovaleric half of the anhydride Uiethylacetyl 85 146-147 iiiolecule enters into an acylation reaction has not Trimethylacetyl 72 144- 145 occurred. This side reaction doubtlessly conIsovderyl 83 1%- 147 tributes to the lowered yields of peptides encounIsocaproyl 36 143- 14B tered in some cases, but has not caused serious Heptanoyl 31h 139- 14I difficulty in the purification of the products obLauroyl 19 143- 144 tained. Purification was effected by recrystallizaHexahydrobenzoyl 37 146-147 tion from ethyl acetate petroleum ether or alcoholPhenylacetyl 1u 142- 141 water mixtures. Hexahydrophenylace tyl ,'6 144-146 Trichloroacetyl 4% 143-141 Experimental'

0

~

O C

Benzoyl 02 141- 11% 46'' 1.7'5- 111 I uroyl 2-Methoxybenzoyl Ih 14:3 144 I-Methoxybenzoy 1 39* < 140 2,4-Dimethoxybenzoy1 38 146-147 .j,1,5-Trimethoxybenzoyl leih 139-141 2-Methylbenzoyl 38 140-142 2-Chlorobenzoyi 34 143- 145 4-Chlorobenzoyl 63* 139- 143 2,4-Dichlorobenzoyl 49 < 140 3,4Dichlorobenzoyl 62 < 110 3-Broinobenzoyl 48* < 14C 2-Iodobenzoyl 43 741- 148 The literature in 11 is 144" (ref 1 ) * These products 11ere still impure after 2- 3 recrystallizations from methanol11

.rtei

I n agreement with the theory proposed, the aand @-branchedchain aliphatic acids, in which both the existing steric effectsand the positive inductive effects of the alky€ groups reinforce each other in the desired manner, were found to be most satisfactory in increasing the acylation ratio of the mixed anhydride in favor of the aminoacid portion of the molecule. The aromatic acids, on the other hand, in which there exists an opposite electronic

Carbobenzox glycinanilide.-A solution of 5.23 g. (0.025 mole) of carbo$nzoxyglycine and 2.55 g. (0.025 mole) of triethylamine in 50 cc of toluene was cooled to 0" and 3.01 g. (0.025 mole) of isovaleryl chioride added. After 2 hours at this temperature, during which time triethylamine hydrochloride separated, 2.34 g. (0.026 mole) of aniline was added. The reaction mixture was then stored a t 8' overnight. The carbobenzoxyglycinanilide crystallized from the reaction mixture and was filtered off together with triethylamine hydrochloride, washed with watert dilute sodium hydroxide solution and dilute hydrochloric acid, and dried; wt. 6.87 g. .(?% yield); A.p. 146-147'. These same reaction conditions were followed exactly uskg the acid chlorides listed in Table I. With fhehigher moLecular xreight aliphatic acid chlorides, and with most of the aromatic acid chlorides, the product separated initially in an impure state and was then recrystallized from aqueous methanol until the melting p&it was above 140". Preparation of Peptide Esters.-In general, the reaction conditions were essentially the same as those used in the preparation of carbobenzoxyglycinanilide. Two examples are given to illustrate minm variations. All solvents and reagents were carefully hied and purikd, prwauti0as were taken to exclude moisture from tbc reaction mixtures. Ethyl Caibobenzo~glycyl-oL-phenqilalanInate .-A solution of 0.025 mole of c a ~ o b e n z o x y g l y c v a ~ ~acid i c anhydride in'Fro cc. of toluene was prepared as described above :11id 1.80 g. (0.025 mole) of ethyl DL-phenylalaninate was arided After storing overnight dt So, the reaction mixturc -

.

4

.I11

i i i t i ~ i i i gp o i n l i

u e iorrctticl.

Dec. 1951

SYNTHESIS OF 2-(3’,4’,5’-TR1METHOXYBENZOYL)-PIPERONYLIC ACID

5555

TABLE I1 PREPARATION O F x-SUBSTITUTED PEPTIDE

ESTERS Analyses,#

Yield, M.P., OC. % (cor..) Ethyl carbohenzoxyglycyl-~~-phenylalaninate” 86 91-92 Ethyl carbobenzoxyglycyl-L-tyrosinate) 77 125-127 Ethyl carbobenzoxy-L-leucylglycinateC 70 103-104 Ethyl dicarbobenzoxy-L-lysylglycinated 58 89-91 Methyl carbobenzoxy-L-leucyl-L-leucinated 29 97-98.5 Ethyl carbobenzoxy-L-leucyl-L-tyrosinate 52 115-117 Ethyl carbobenzoxy-L-leucyl-oL-phenylalaninate 38 84-87 Ethyl carbobenzoxy-DL-alanyl-DL-phenylalaninate 48 114-116 Ethyl carbobenzoxyglycyl-DL-phenylalanylglycinatef 86 132-133 Ethyl phthalylglycyl-DL-phenylalaninate 68 149-150 Ethyl phthalylglycyl-L-tyrosinate 60 163-164 Ethyl phthalylglycyl-L-leucinate 62 139-140 Ethyl phthalyl-DL-alanyl-DL-valinate 40 110-130

[ulrrD

(ethanol)

Formula

C

Calcd. H

N

% -- Found

C

H

N

. . . . . . . . . ... 4-18.4 (C = -25.6 (C - 1 2 . 0 ( ~= - 3 5 . 3 (C = - 1 5 . 2 (C =

5) 6)

5) 10) 5) C Z L H ~ ~ N ~65.77 O O 7.07 6 . 1 4 65.68 7.22 6.28 . . . . . . . . . . . CzaHs?NtOs 6 8 . 1 6 7 . 1 6 6 . 3 6 7 . 2 5 6.24 . . . . . . . . . . . C ~ H Z ~ N Z O66.31 I 6 . 5 8 7 . 0 3 66.22 6 . 6 5 7 . 1 7

.

.

............ ,........... + 4 3 . 0 (C = 2) - 2 4 . 5 (C = 2)

............

CzaHnrN~Os Cx~HxsNzOr CziHxsNzOs CisHxnNxOr CuHxrN2Or

62.57 66.30 63.63 62.41 62.41

6.17 5.30 5.05 6.40 6.40

9.52 7.37 7.07 8.09 8.09

62.32 66.28 63.40 62.20 62.34

6.24 6.36 5.10 6.44 6.56

9.64 7.43 7.27 8.05 8.26

H. Keurath, et al., J . Biol. Chem., 170, 221 (1947),give m.p. 90-91’ (cor.). M. Bergm2nn and J. S. Fruton, ibid., 118,405(1937),give m.p. 118”. M. Bergmann, et al., ibid., 111,225(1935),give m.p. 103-104 ; M. A. Nyman and R. M. Herbst, J . Org. Chem., 15 108 (1950),give m.p. 99’ and [ C Y ] ~ D-26.8’ (c = 2.6ethanol). M. Bergmann, el al., 2. physiol. Chem., 224,26 (1934),give m.p. 90”. e M. A. Nyman and R. M. Herbst, ibid., give m.p. 97-98’. Prepared from carbobenzoxyglycyl-m-phenylalanineand ethyl glycinate. We are indebted t o Dr. J. A. Kuck and his staff of these laboratories for the microanalyses. The values reported are the average of two values differing by not more than 0.30. A satisfactory a

carbon value was not obtained on this compound. was washed with water and the product which had crystallized out was filtered off and washed with 3% sodium bicarbonate solution; wt. 5.29 g. (55% yield), m.p. 91-92’. The toluene layer was separated from the filtrate, washed rapidly with 3% sodium bicarbonate solution and diluted until cloudy with petroleum ether and cooled to crystallize a second crop of material; wt. 3.14 g. (33% yield); m.p. 90-91’. The two crops were combined and recrystallized from aqueous ethanol to give 8.25 g. (86%) of product melting a t 91-92’, Ethyl Carbobenzoxy-L-1eucylglycinate.-A solution of 5.30 g. (0.02mole) of carbobenzoxy-L-leucine and 2.04 g. (0.02 mole) of triethylamine in a mixture of 25 cc. of toluene and 25 cc. of chloroform was cooled to -5’ and 2.41 g. (0.02 mole) of isovaleryl chloride added. -4fter 1.5hours a second,

[CONTRIBUTION FROM

THE

precooled solution of 2.79 g. (0.02mole) of ethyl glycinate hydrochloride and 2.04 g. (0.02 mole) of triethylamine in 50 cc. of chloroform was added, and the reaction mixture was stored overnight at 8 ” . The organic solution was washed with water and with 3% sodium bicarbonate solution and diluted until cloudy with petroleum ether. On cooling, the product separated as colorless crystals; wt. 2.17g.(31%), m.p. 103-104’. Concentration of the mother liquor almost to dryness in a stream of air and rediluting with petroleum ether gave a second crop of crystalline product; wt. 2.98 g. (43%); m.p. 102-103’. The two crops were combined and recrystallized from aqueous ethanol t o give 4.90 g. (70%) of pure product melting a t 105-106”. STAMFORD, CONNECTICUT RECEIVED MAY31, 1951

CHEMISTRY DEPARTMENT O F BOSTON UNIVERSITY]

Synthesis of 2-(3 ’,4’,5’-Trimethoxybenzoyl)-piperonylicAcid1i2 BY WALTERJ, GENSLERAND CARLOSM. SAMOUR A synthesis for 2-(3’,4’,5’-trimethoxybenzoy1)-piperonylic acid is described according to the following sequence: homopiperonylamine, N-(trimethoxybenzoy1)-homopiperonylamine, 1-(3’,4’,5’-trimethoxyphenyl)-6,7-methylenedioxy-3,4-dihydroisoquinoline, 2-(3’,4’,5’-trimethoxybenzoyl)-4,5-methylenedioxystyrene and 2-(3’,4’,5’-trirnethoxybenzoy1)-piperonylic acid. The over-all yield for the four steps is 60%.

I n view of the possibility of elaborating picropodophyllin (11) or the related podophyllotoxin3 from 2-(3’,4’,5’-trimethoxybenzoyl)-piperonylic acid (I), a convenient source of this keto acid would be desirable. Although 2-(3’,4’,5’-trimethoxybenzoy1)-piperonylic acid (I) had been obtained before, not only from picropodophyllin (11)4itself, but also from a synthetic dihydronaphthalene derivative5 and from several derivatives of 1-(3’,4’,5’-trimeth(1) This work was supported by American Cancer Society Grant-inAid No. CBC-6 as recommended by the Committee on Growth of the National Research Council. (2) A summary of the material in this paper was presented in Boston, Mass., on April 3, 1951, before the Division of Organic Chemistry of the American Chemical Society. (3) Podophyllotoxin, a tumor-damaging substance, is currently of interest a s a potentially useful anti-cancer agent. For example, see the report by J. Leiter, V. Downing, J. L. Hartwell and M. J. Shear, J . N a f . Cancer I n s f . , 10, 1273 (1950). (4) E. Sphth, P. Wessely and E. Nadler, Ber., 66, 125 (1933). (5) R . D. Haworth and T. Richardson, J. Ckem. SOC.,348 (1936). See also R. D.Haworth and J . R . Atkinson, i b i d . , 797 (1938).

OH

oxyphenyl) -6,7-methylenedio~yiquinoline,~ none of the paths constituted a n attractive preparative method. We wish now to report on a practical synthesis of keto-acid (I).’ (6) W. Reeve and W, M. Eareckson, 111, TRISJOURNAL, 73, 5195 11954).

(7) A brief account of this work has appeared, ibid., 79, 3318 (1950). In the preliminary report, reference to the work of Haworth and Richardson (ref. 5) was inadvertently omitted.